Reliability and validity of a transdiagnostic measure of reward valuation effort.

Abstract

To identify biobehavioral mechanisms underlying excessive reward consumption, reward valuation—effort (RV-E) assessments should (a) parallel measures in basic science to permit translation from preclinical to clinical studies; (b) quantify constructs dimensionally from healthy to disease states; and (c) hold relevance across different diagnostic categories. To address these aims, we developed a progressive ratio (PR) task whereby RV-E is measured as breakpoint when participants worked for access to playing a game. We evaluated test–retest reliability of breakpoint and convergent and discriminant validity of interpretations of this score against an established PR task for food. In Study 1, female undergraduates (N = 71; 33% racial minority; 28% ethnic minority) completed the game and food tasks in fasted and fed states. In Study 2, women (N = 189; 29% racial minority; 27% ethnic minority) with eating disorders (n = 158) were compared to controls (n = 31) on tasks. Game task breakpoint demonstrated excellent test–retest reliability, intraclass correlation coefficient (ICC) = .91, 95% CI [.80, −.96], over 2 weeks and convergent validity with the fasted food task (r = .51, p < .001). Consistent with animal models, breakpoint was lower in fed compared to fasted states across tasks, B (SE) = 321.01 (552.40), p < .001. Finally, the game task demonstrated discriminant validity from measurement of satiation. In Study 2, women with eating disorders demonstrated higher breakpoint on both tasks compared to controls, and game PR task breakpoint decreased from a fasted to fed state. The game PR task offers a novel approach for translating results from animal models of RV-E into testable hypotheses in nonclinical and clinical samples.

Greater reliance on model-free learning in adolescent anorexia nervosa: An examination of dual-system reinforcement learning

“…If individuals with AN-BP are indeed more sensitive to reward, it is possible that positive outcomes (e.g., monetary gain, palatable food consumption) have higher motivational and incentive salience, leading to use of a cognitively demanding strategy even in the context of lower trait self-regulatory control. This is consistent with findings that adults with bulimic-type symptoms are more willing to work for reward as demonstrated by a higher breakpoint during a progressive ratio task 86 , and prior research demonstrating that model-based learning can be modulated by stress and cognitive control 61,87 .…”

A mechanistic staging model of reward processing alterations in individuals with binge‐type eating disorders

“…Moreover, in support of our model suggesting an association between ‘wanting’ and duration of binge eating, higher breakpoint on the food PR task was associated with longer duration of illness in women with BN ( r = .34). An identical nonfood PR task, in which participants work to earn play time of the game Angry Birds, also has been developed (Keel et al, 2022). With the establishment of both food and nonfood reward tasks, we can examine changes in ‘wanting’ for food versus other primary reinforces with illness stage.…”